Kedar Sathe
// proof-of-work

proof-of-work

building, shipping, and learning in public

Loading projects...

Detailed Project Descriptions

LazyDocs 📚

LazyDocs (Live Demo) is a CLI-based NPM module that automates documentation generation for your codebase using the GroQ API. Built as a Node.js command-line tool, LazyDocs analyzes your project structure and generates comprehensive, well-structured documentation with a simple terminal command. Powered by GroQ's fast inference capabilities, it understands code context, identifies key components, and produces clear, maintainable documentation instantly. The tool is designed for seamless integration into any development workflow - just install via npm and run the CLI command to generate docs. Perfect for developers who want high-quality documentation without the manual overhead, LazyDocs brings AI-powered documentation generation directly to your terminal.

Pelvix AI 🤖

Pelvix AI (Live Demo) is your digital best friend powered by AI - a modern, full-stack chatbot with personality built with Next.js 15 and React 19. Features include intelligent chat powered by DeepSeek-R1 via Pollinations AI, high-quality image generation with /imagine commands, smooth GSAP-powered animations, and a warm, friendly personality. The application includes real-time chat functionality, custom dark themes, robust error handling, and is optimized for performance and responsiveness across all devices.

X-Bird - AI Tweet Reply Generator

X-Bird is a Chrome extension that generates smart, contextual tweet replies on X.com using AI. Powered by Hugging Face's Qwen2.5-72B-Instruct model (Alibaba's 72B parameter model), it features multiple mood options (Auto, Savage, Supportive, Funny, Professional, Casual), adjustable brainrot intensity control (0-100), intelligent tweet filtering for high-engagement content, secure API key storage, and seamless UI integration with Twitter's interface.

RAG Application System

RAG Application is a comprehensive Retrieval-Augmented Generation system that combines the power of large language models with external knowledge retrieval. This system enhances AI responses by incorporating relevant information from external databases, significantly improving accuracy and reducing hallucinations. The architecture includes vector embeddings, semantic search capabilities, and intelligent knowledge integration for contextually relevant responses.

Snake's & The Golden Apple

Snake's & The Golden Apple is a competitive, AI-driven game where two snakes learn to navigate, collect food, and avoid collisions using Deep Q-Learning. This project demonstrates reinforcement learning in a dynamic environment, showcasing how agents evolve their strategies over time.

Cybur

Cybur is a Loadable Kernel Module (LKM) designed to detect and prevent ARP Poisoning Attacks. It intercepts ARP packets, validates their authenticity, and cross-references them with a dynamically maintained DHCP snooping table to secure the ARP cache.

FreeLix

FreeLix is an open-source web application built with React and Tailwind CSS that democratizes AI-powered language processing by providing free transcription and translation services. The application leverages the powerful Whisper-large-v3 model from Hugging Face to convert audio to text with high accuracy, supporting multiple languages and formats. FreeLix also offers multilingual text translation capabilities, making it a comprehensive solution for breaking down language barriers. The clean, responsive interface ensures accessibility across all devices, while the open-source nature allows developers to contribute and customize the platform. Perfect for students, content creators, researchers, and anyone needing reliable transcription and translation services without cost barriers.

Image Super-Resolution GAN

Image Super-Resolution GAN employs a Super Resolution Generative Adversarial Network to convert low-resolution images into high-quality, photo-realistic outputs.

Neural Black - Brain Tumor Diagnosis System

Neural Black - Brain Tumor Diagnosis System is an end-to-end solution for detecting, classifying, and diagnosing brain tumors. It leverages advanced Deep Learning methods to achieve high diagnostic accuracy.

Third Eye - Intrusion Detection System

Third Eye - Intrusion Detection System integrates a firewall with Machine Learning algorithms to analyze network traffic and detect potential intrusions, enhancing overall network security.

Malware Prediction

Malware Prediction utilizes Machine Learning techniques to assess portable executable applications, predicting whether they are malware or at risk of being attacked.